The Agent Skills Directory

[PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection due to its ability to process untrusted data from an external task management service while maintaining powerful action capabilities.
Ingestion points: External data enters the agent context through task retrieval tools such as mcp__dida__get_task_by_id, mcp__dida__search_task, and mcp__dida__list_undone_tasks_by_date as defined in SKILL.md.
Boundary markers: Absent. The instructions do not specify delimiters or provide warnings to the agent to disregard instructions embedded within task data.
Capability inventory: The skill includes a wide range of destructive and creative capabilities, including create_task, update_task, batch_update_tasks, and complete_task, which could be exploited if an injection is successful.
Sanitization: Absent. There is no evidence of content validation or sanitization for the data retrieved from the Dida365 API before it is processed by the agent.

dida365